Michael Rogers wrote:
John Bäckstrand wrote:
I have kind of figured this out from the messages from CIA, that
simulator will be very low-level. I am a bit concerned though.. wont
this make it extremely slow? Does it have any way of falling back to
somewhat simpler and faster models? If its faster to actually deploy a
testnet than to simulate you don't actually gain anything ;)
Once the lower levels are behaving as expected, I'll implement a 'simple
mode' that just delivers messages directly from one node to the next
without simulating the low-level transport. But the 'detailed mode'
should always be there as a sanity check, because there can be
unexpected interactions between layers (especially where things like
timeouts are concerned).
Yeah, sounds good. I'm officially not concerned anymore ;) Nice work and
good luck.
---
John Bäckstrand
_______________________________________________
Devl mailing list
Devl@freenetproject.org
http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
